r303031 - [DOXYGEN] Minor improvements in doxygen comments.

Ekaterina Romanova via cfe-commits cfe-commits at lists.llvm.org
Sun May 14 20:25:05 PDT 2017

Author: kromanova
Date: Sun May 14 22:25:04 2017
New Revision: 303031

URL: http://llvm.org/viewvc/llvm-project?rev=303031&view=rev
Log:
[DOXYGEN] Minor improvements in doxygen comments.

Separated very long brief sections into two sections.

I got an OK from Eric Christopher to commit doxygen comments without prior code
review upstream.

Modified:

==============================================================================
+++ cfe/trunk/lib/Headers/avxintrin.h Sun May 14 22:25:04 2017
@@ -1458,12 +1458,13 @@ _mm256_blendv_ps(__m256 __a, __m256 __b,
/// \brief Computes two dot products in parallel, using the lower and upper
///    halves of two [8 x float] vectors as input to the two computations, and
///    returning the two dot products in the lower and upper halves of the
-///    [8 x float] result. The immediate integer operand controls which input
-///    elements will contribute to the dot product, and where the final results
-///    are returned. In general, for each dot product, the four corresponding
-///    elements of the input vectors are multiplied; the first two and second
-///    two products are summed, then the two sums are added to form the final
-///    result.
+///    [8 x float] result.
+///
+///    The immediate integer operand controls which input elements will
+///    contribute to the dot product, and where the final results are returned.
+///    In general, for each dot product, the four corresponding elements of the
+///    input vectors are multiplied; the first two and second two products are
+///    summed, then the two sums are added to form the final result.
///
///
@@ -1497,15 +1498,16 @@ _mm256_blendv_ps(__m256 __a, __m256 __b,

/* Vector shuffle */
/// \brief Selects 8 float values from the 256-bit operands of [8 x float], as
-///    specified by the immediate value operand. The four selected elements in
-///    each operand are copied to the destination according to the bits
-///    specified in the immediate operand. The selected elements from the first
-///    256-bit operand are copied to bits [63:0] and bits [191:128] of the
-///    destination, and the selected elements from the second 256-bit operand
-///    are copied to bits [127:64] and bits [255:192] of the destination. For
-///    example, if bits [7:0] of the immediate operand contain a value of 0xFF,
-///    the 256-bit destination vector would contain the following values: b,
-///    b, a, a, b, b, a, a.
+///    specified by the immediate value operand.
+///
+///    The four selected elements in each operand are copied to the destination
+///    according to the bits specified in the immediate operand. The selected
+///    elements from the first 256-bit operand are copied to bits [63:0] and
+///    bits [191:128] of the destination, and the selected elements from the
+///    second 256-bit operand are copied to bits [127:64] and bits [255:192] of
+///    the destination. For example, if bits [7:0] of the immediate operand
+///    contain a value of 0xFF, the 256-bit destination vector would contain the
+///    following values: b, b, a, a, b, b, a, a.
///
///
@@ -1557,13 +1559,14 @@ _mm256_blendv_ps(__m256 __a, __m256 __b,
12 + (((mask) >> 6) & 0x3)); })

/// \brief Selects four double-precision values from the 256-bit operands of
-///    [4 x double], as specified by the immediate value operand. The selected
-///    elements from the first 256-bit operand are copied to bits [63:0] and
-///    bits [191:128] in the destination, and the selected elements from the
-///    second 256-bit operand are copied to bits [127:64] and bits [255:192] in
-///    the destination. For example, if bits [3:0] of the immediate operand
-///    contain a value of 0xF, the 256-bit destination vector would contain the
-///    following values: b, a, b, a.
+///    [4 x double], as specified by the immediate value operand.
+///
+///    The selected elements from the first 256-bit operand are copied to bits
+///    [63:0] and bits [191:128] in the destination, and the selected elements
+///    from the second 256-bit operand are copied to bits [127:64] and bits
+///    [255:192] in the destination. For example, if bits [3:0] of the immediate
+///    operand contain a value of 0xF, the 256-bit destination vector would
+///    contain the following values: b, a, b, a.
///
///
@@ -1641,9 +1644,11 @@ _mm256_blendv_ps(__m256 __a, __m256 __b,

/// \brief Compares each of the corresponding double-precision values of two
///    128-bit vectors of [2 x double], using the operation specified by the
-///    immediate integer operand. Returns a [2 x double] vector consisting of
-///    two doubles corresponding to the two comparison results: zero if the
-///    comparison is false, and all 1's if the comparison is true.
+///    immediate integer operand.
+///
+///    Returns a [2 x double] vector consisting of two doubles corresponding to
+///    the two comparison results: zero if the comparison is false, and all 1's
+///    if the comparison is true.
///
///
@@ -1699,9 +1704,11 @@ _mm256_blendv_ps(__m256 __a, __m256 __b,

/// \brief Compares each of the corresponding values of two 128-bit vectors of
///    [4 x float], using the operation specified by the immediate integer
-///    operand. Returns a [4 x float] vector consisting of four floats
-///    corresponding to the four comparison results: zero if the comparison is
-///    false, and all 1's if the comparison is true.
+///    operand.
+///
+///    Returns a [4 x float] vector consisting of four floats corresponding to
+///    the four comparison results: zero if the comparison is false, and all 1's
+///    if the comparison is true.
///
///
@@ -1757,9 +1764,11 @@ _mm256_blendv_ps(__m256 __a, __m256 __b,

/// \brief Compares each of the corresponding double-precision values of two
///    256-bit vectors of [4 x double], using the operation specified by the
-///    immediate integer operand. Returns a [4 x double] vector consisting of
-///    four doubles corresponding to the four comparison results: zero if the
-///    comparison is false, and all 1's if the comparison is true.
+///    immediate integer operand.
+///
+///    Returns a [4 x double] vector consisting of four doubles corresponding to
+///    the four comparison results: zero if the comparison is false, and all 1's
+///    if the comparison is true.
///
///
@@ -1815,9 +1824,11 @@ _mm256_blendv_ps(__m256 __a, __m256 __b,

/// \brief Compares each of the corresponding values of two 256-bit vectors of
///    [8 x float], using the operation specified by the immediate integer
-///    operand. Returns a [8 x float] vector consisting of eight floats
-///    corresponding to the eight comparison results: zero if the comparison is
-///    false, and all 1's if the comparison is true.
+///    operand.
+///
+///    Returns a [8 x float] vector consisting of eight floats corresponding to
+///    the eight comparison results: zero if the comparison is false, and all
+///    1's if the comparison is true.
///
///
@@ -1873,8 +1884,10 @@ _mm256_blendv_ps(__m256 __a, __m256 __b,

/// \brief Compares each of the corresponding scalar double-precision values of
///    two 128-bit vectors of [2 x double], using the operation specified by the
-///    immediate integer operand. If the result is true, all 64 bits of the
-///    destination vector are set; otherwise they are cleared.
+///    immediate integer operand.
+///
+///    If the result is true, all 64 bits of the destination vector are set;
+///    otherwise they are cleared.
///
///
@@ -1930,8 +1943,10 @@ _mm256_blendv_ps(__m256 __a, __m256 __b,

/// \brief Compares each of the corresponding scalar values of two 128-bit
///    vectors of [4 x float], using the operation specified by the immediate
-///    integer operand. If the result is true, all 32 bits of the destination
-///    vector are set; otherwise they are cleared.
+///    integer operand.
+///
+///    If the result is true, all 32 bits of the destination vector are set;
+///    otherwise they are cleared.
///
///
@@ -2536,7 +2551,9 @@ _mm256_unpacklo_ps(__m256 __a, __m256 __
/// \brief Given two 128-bit floating-point vectors of [2 x double], perform an
///    element-by-element comparison of the double-precision element in the
///    first source vector and the corresponding element in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of double-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2563,7 +2580,9 @@ _mm_testz_pd(__m128d __a, __m128d __b)
/// \brief Given two 128-bit floating-point vectors of [2 x double], perform an
///    element-by-element comparison of the double-precision element in the
///    first source vector and the corresponding element in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of double-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2590,7 +2609,9 @@ _mm_testc_pd(__m128d __a, __m128d __b)
/// \brief Given two 128-bit floating-point vectors of [2 x double], perform an
///    element-by-element comparison of the double-precision element in the
///    first source vector and the corresponding element in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of double-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2618,7 +2639,9 @@ _mm_testnzc_pd(__m128d __a, __m128d __b)
/// \brief Given two 128-bit floating-point vectors of [4 x float], perform an
///    element-by-element comparison of the single-precision element in the
///    first source vector and the corresponding element in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of single-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2645,7 +2668,9 @@ _mm_testz_ps(__m128 __a, __m128 __b)
/// \brief Given two 128-bit floating-point vectors of [4 x float], perform an
///    element-by-element comparison of the single-precision element in the
///    first source vector and the corresponding element in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of single-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2672,7 +2697,9 @@ _mm_testc_ps(__m128 __a, __m128 __b)
/// \brief Given two 128-bit floating-point vectors of [4 x float], perform an
///    element-by-element comparison of the single-precision element in the
///    first source vector and the corresponding element in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of single-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2700,7 +2727,9 @@ _mm_testnzc_ps(__m128 __a, __m128 __b)
/// \brief Given two 256-bit floating-point vectors of [4 x double], perform an
///    element-by-element comparison of the double-precision elements in the
///    first source vector and the corresponding elements in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of double-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2727,7 +2756,9 @@ _mm256_testz_pd(__m256d __a, __m256d __b
/// \brief Given two 256-bit floating-point vectors of [4 x double], perform an
///    element-by-element comparison of the double-precision elements in the
///    first source vector and the corresponding elements in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of double-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2754,7 +2785,9 @@ _mm256_testc_pd(__m256d __a, __m256d __b
/// \brief Given two 256-bit floating-point vectors of [4 x double], perform an
///    element-by-element comparison of the double-precision elements in the
///    first source vector and the corresponding elements in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of double-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2782,7 +2815,9 @@ _mm256_testnzc_pd(__m256d __a, __m256d _
/// \brief Given two 256-bit floating-point vectors of [8 x float], perform an
///    element-by-element comparison of the single-precision element in the
///    first source vector and the corresponding element in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of single-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2809,7 +2844,9 @@ _mm256_testz_ps(__m256 __a, __m256 __b)
/// \brief Given two 256-bit floating-point vectors of [8 x float], perform an
///    element-by-element comparison of the single-precision element in the
///    first source vector and the corresponding element in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of single-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2836,7 +2873,9 @@ _mm256_testc_ps(__m256 __a, __m256 __b)
/// \brief Given two 256-bit floating-point vectors of [8 x float], perform an
///    element-by-element comparison of the single-precision elements in the
///    first source vector and the corresponding elements in the second source
-///    vector. The EFLAGS register is updated as follows: \n
+///    vector.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of single-precision elements where the
///    sign-bits of both elements are 1, the ZF flag is set to 0. Otherwise the
///    ZF flag is set to 1. \n
@@ -2862,7 +2901,9 @@ _mm256_testnzc_ps(__m256 __a, __m256 __b
}

/// \brief Given two 256-bit integer vectors, perform a bit-by-bit comparison
-///    of the two source vectors and update the EFLAGS register as follows: \n
+///    of the two source vectors.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of bits where both bits are 1, the ZF flag
///    is set to 0. Otherwise the ZF flag is set to 1. \n
///    If there is at least one pair of bits where the bit from the first source
@@ -2886,7 +2927,9 @@ _mm256_testz_si256(__m256i __a, __m256i
}

/// \brief Given two 256-bit integer vectors, perform a bit-by-bit comparison
-///    of the two source vectors and update the EFLAGS register as follows: \n
+///    of the two source vectors.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of bits where both bits are 1, the ZF flag
///    is set to 0. Otherwise the ZF flag is set to 1. \n
///    If there is at least one pair of bits where the bit from the first source
@@ -2910,7 +2953,9 @@ _mm256_testc_si256(__m256i __a, __m256i
}

/// \brief Given two 256-bit integer vectors, perform a bit-by-bit comparison
-///    of the two source vectors and update the EFLAGS register as follows: \n
+///    of the two source vectors.
+///
+///    The EFLAGS register is updated as follows: \n
///    If there is at least one pair of bits where both bits are 1, the ZF flag
///    is set to 0. Otherwise the ZF flag is set to 1. \n
///    If there is at least one pair of bits where the bit from the first source
@@ -4466,9 +4511,10 @@ _mm256_castsi256_si128(__m256i __a)
}

/// \brief Constructs a 256-bit floating-point vector of [4 x double] from a
-///    128-bit floating-point vector of [2 x double]. The lower 128 bits
-///    contain the value of the source vector. The contents of the upper 128
-///    bits are undefined.
+///    128-bit floating-point vector of [2 x double].
+///
+///    The lower 128 bits contain the value of the source vector. The contents
+///    of the upper 128 bits are undefined.
///
///
@@ -4486,9 +4532,10 @@ _mm256_castpd128_pd256(__m128d __a)
}

/// \brief Constructs a 256-bit floating-point vector of [8 x float] from a
-///    128-bit floating-point vector of [4 x float]. The lower 128 bits contain
-///    the value of the source vector. The contents of the upper 128 bits are
-///    undefined.
+///    128-bit floating-point vector of [4 x float].
+///
+///    The lower 128 bits contain the value of the source vector. The contents
+///    of the upper 128 bits are undefined.
///
///
@@ -4506,6 +4553,7 @@ _mm256_castps128_ps256(__m128 __a)
}

/// \brief Constructs a 256-bit integer vector from a 128-bit integer vector.
+///
///    The lower 128 bits contain the value of the source vector. The contents
///    of the upper 128 bits are undefined.
///
@@ -4586,8 +4634,10 @@ _mm256_zextsi128_si256(__m128i __a)
/// \brief Constructs a new 256-bit vector of [8 x float] by first duplicating
///    a 256-bit vector of [8 x float] given in the first parameter, and then
///    replacing either the upper or the lower 128 bits with the contents of a
-///    128-bit vector of [4 x float] in the second parameter. The immediate
-///    integer parameter determines between the upper or the lower 128 bits.
+///    128-bit vector of [4 x float] in the second parameter.
+///
+///    The immediate integer parameter determines between the upper or the lower
+///    128 bits.
///
///
@@ -4631,8 +4681,10 @@ _mm256_zextsi128_si256(__m128i __a)
/// \brief Constructs a new 256-bit vector of [4 x double] by first duplicating
///    a 256-bit vector of [4 x double] given in the first parameter, and then
///    replacing either the upper or the lower 128 bits with the contents of a
-///    128-bit vector of [2 x double] in the second parameter. The immediate
-///    integer parameter determines between the upper or the lower 128 bits.
+///    128-bit vector of [2 x double] in the second parameter.
+///
+///    The immediate integer parameter determines between the upper or the lower
+///    128 bits.
///
///
@@ -4672,8 +4724,10 @@ _mm256_zextsi128_si256(__m128i __a)
/// \brief Constructs a new 256-bit integer vector by first duplicating a
///    256-bit integer vector given in the first parameter, and then replacing
///    either the upper or the lower 128 bits with the contents of a 128-bit
-///    integer vector in the second parameter. The immediate integer parameter
-///    determines between the upper or the lower 128 bits.
+///    integer vector in the second parameter.
+///
+///    The immediate integer parameter determines between the upper or the lower
+///    128 bits.
///
///

==============================================================================
+++ cfe/trunk/lib/Headers/emmintrin.h Sun May 14 22:25:04 2017
@@ -462,8 +462,9 @@ _mm_cmplt_pd(__m128d __a, __m128d __b)

/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
-///    operand are less than or equal to those in the second operand. Each
-///    comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+///    operand are less than or equal to those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -482,8 +483,9 @@ _mm_cmple_pd(__m128d __a, __m128d __b)

/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
-///    operand are greater than those in the second operand. Each comparison
-///    yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+///    operand are greater than those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -502,8 +504,9 @@ _mm_cmpgt_pd(__m128d __a, __m128d __b)

/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
-///    operand are greater than or equal to those in the second operand. Each
-///    comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+///    operand are greater than or equal to those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -522,9 +525,10 @@ _mm_cmpge_pd(__m128d __a, __m128d __b)

/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
-///    operand are ordered with respect to those in the second operand. A pair
-///    of double-precision values are "ordered" with respect to each other if
-///    neither value is a NaN. Each comparison yields 0h for false,
+///    operand are ordered with respect to those in the second operand.
+///
+///    A pair of double-precision values are "ordered" with respect to each
+///    other if neither value is a NaN. Each comparison yields 0h for false,
///    FFFFFFFFFFFFFFFFh for true.
///
@@ -544,9 +548,10 @@ _mm_cmpord_pd(__m128d __a, __m128d __b)

/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
-///    operand are unordered with respect to those in the second operand. A pair
-///    of double-precision values are "unordered" with respect to each other if
-///    one or both values are NaN. Each comparison yields 0h for false,
+///    operand are unordered with respect to those in the second operand.
+///
+///    A pair of double-precision values are "unordered" with respect to each
+///    other if one or both values are NaN. Each comparison yields 0h for false,
///    FFFFFFFFFFFFFFFFh for true.
///
@@ -567,8 +572,9 @@ _mm_cmpunord_pd(__m128d __a, __m128d __b

/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
-///    operand are unequal to those in the second operand. Each comparison
-///    yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+///    operand are unequal to those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -587,8 +593,9 @@ _mm_cmpneq_pd(__m128d __a, __m128d __b)

/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
-///    operand are not less than those in the second operand. Each comparison
-///    yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+///    operand are not less than those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -607,8 +614,9 @@ _mm_cmpnlt_pd(__m128d __a, __m128d __b)

/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
-///    operand are not less than or equal to those in the second operand. Each
-///    comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+///    operand are not less than or equal to those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -627,8 +635,9 @@ _mm_cmpnle_pd(__m128d __a, __m128d __b)

/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
-///    operand are not greater than those in the second operand. Each
-///    comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+///    operand are not greater than those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -648,6 +657,7 @@ _mm_cmpngt_pd(__m128d __a, __m128d __b)
/// \brief Compares each of the corresponding double-precision values of the
///    128-bit vectors of [2 x double] to determine if the values in the first
///    operand are not greater than or equal to those in the second operand.
+///
///    Each comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
@@ -666,8 +676,9 @@ _mm_cmpnge_pd(__m128d __a, __m128d __b)
}

/// \brief Compares the lower double-precision floating-point values in each of
-///    the two 128-bit floating-point vectors of [2 x double] for equality. The
-///    comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
+///    the two 128-bit floating-point vectors of [2 x double] for equality.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -690,8 +701,9 @@ _mm_cmpeq_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is less than the corresponding value in
-///    the second parameter. The comparison yields 0h for false,
-///    FFFFFFFFFFFFFFFFh for true.
+///    the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -714,8 +726,9 @@ _mm_cmplt_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is less than or equal to the
-///    corresponding value in the second parameter. The comparison yields 0h for
-///    false, FFFFFFFFFFFFFFFFh for true.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -738,8 +751,9 @@ _mm_cmple_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is greater than the corresponding value
-///    in the second parameter. The comparison yields 0h for false,
-///    FFFFFFFFFFFFFFFFh for true.
+///    in the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -763,8 +777,9 @@ _mm_cmpgt_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is greater than or equal to the
-///    corresponding value in the second parameter. The comparison yields 0h for
-///    false, FFFFFFFFFFFFFFFFh for true.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -788,9 +803,11 @@ _mm_cmpge_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is "ordered" with respect to the
-///    corresponding value in the second parameter. The comparison yields 0h for
-///    false, FFFFFFFFFFFFFFFFh for true. A pair of double-precision values are
-///    "ordered" with respect to each other if neither value is a NaN.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. A pair of
+///    double-precision values are "ordered" with respect to each other if
+///    neither value is a NaN.
///
///
@@ -813,9 +830,11 @@ _mm_cmpord_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is "unordered" with respect to the
-///    corresponding value in the second parameter. The comparison yields 0h
-///    for false, FFFFFFFFFFFFFFFFh for true. A pair of double-precision values
-///    are "unordered" with respect to each other if one or both values are NaN.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true. A pair of
+///    double-precision values are "unordered" with respect to each other if one
+///    or both values are NaN.
///
///
@@ -839,8 +858,9 @@ _mm_cmpunord_sd(__m128d __a, __m128d __b
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is unequal to the corresponding value in
-///    the second parameter. The comparison yields 0h for false,
-///    FFFFFFFFFFFFFFFFh for true.
+///    the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -863,8 +883,9 @@ _mm_cmpneq_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is not less than the corresponding
-///    value in the second parameter. The comparison yields 0h for false,
-///    FFFFFFFFFFFFFFFFh for true.
+///    value in the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -887,8 +908,9 @@ _mm_cmpnlt_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is not less than or equal to the
-///    corresponding value in the second parameter. The comparison yields 0h
-///    for false, FFFFFFFFFFFFFFFFh for true.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -911,8 +933,9 @@ _mm_cmpnle_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is not greater than the corresponding
-///    value in the second parameter. The comparison yields 0h for false,
-///    FFFFFFFFFFFFFFFFh for true.
+///    value in the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -936,8 +959,9 @@ _mm_cmpngt_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is not greater than or equal to the
-///    corresponding value in the second parameter. The comparison yields 0h
-///    for false, FFFFFFFFFFFFFFFFh for true.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0h for false, FFFFFFFFFFFFFFFFh for true.
///
///
@@ -982,7 +1006,9 @@ _mm_comieq_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is less than the corresponding value in
-///    the second parameter. The comparison yields 0 for false, 1 for true.
+///    the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true.
///
///
@@ -1004,8 +1030,9 @@ _mm_comilt_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is less than or equal to the
-///    corresponding value in the second parameter. The comparison yields 0 for
-///    false, 1 for true.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true.
///
///
@@ -1027,7 +1054,9 @@ _mm_comile_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is greater than the corresponding value
-///    in the second parameter. The comparison yields 0 for false, 1 for true.
+///    in the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true.
///
///
@@ -1049,8 +1078,9 @@ _mm_comigt_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is greater than or equal to the
-///    corresponding value in the second parameter. The comparison yields 0 for
-///    false, 1 for true.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true.
///
///
@@ -1072,7 +1102,9 @@ _mm_comige_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is unequal to the corresponding value in
-///    the second parameter. The comparison yields 0 for false, 1 for true.
+///    the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true.
///
///
@@ -1093,8 +1125,9 @@ _mm_comineq_sd(__m128d __a, __m128d __b)

/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] for equality. The
-///    comparison yields 0 for false, 1 for true. If either of the two lower
-///    double-precision values is NaN, 1 is returned.
+///    comparison yields 0 for false, 1 for true.
+///
+///    If either of the two lower double-precision values is NaN, 1 is returned.
///
///
@@ -1117,8 +1150,10 @@ _mm_ucomieq_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is less than the corresponding value in
-///    the second parameter. The comparison yields 0 for false, 1 for true. If
-///    either of the two lower double-precision values is NaN, 1 is returned.
+///    the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true. If either of the two lower
+///    double-precision values is NaN, 1 is returned.
///
///
@@ -1141,9 +1176,10 @@ _mm_ucomilt_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is less than or equal to the
-///    corresponding value in the second parameter. The comparison yields 0 for
-///    false, 1 for true. If either of the two lower double-precision values is
-///    NaN, 1 is returned.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true. If either of the two lower
+///    double-precision values is NaN, 1 is returned.
///
///
@@ -1166,8 +1202,10 @@ _mm_ucomile_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is greater than the corresponding value
-///    in the second parameter. The comparison yields 0 for false, 1 for true.
-///    If either of the two lower double-precision values is NaN, 0 is returned.
+///    in the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true. If either of the two lower
+///    double-precision values is NaN, 0 is returned.
///
///
@@ -1190,9 +1228,10 @@ _mm_ucomigt_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is greater than or equal to the
-///    corresponding value in the second parameter. The comparison yields 0 for
-///    false, 1 for true.  If either of the two lower double-precision values
-///    is NaN, 0 is returned.
+///    corresponding value in the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true.  If either of the two
+///    lower double-precision values is NaN, 0 is returned.
///
///
@@ -1215,8 +1254,10 @@ _mm_ucomige_sd(__m128d __a, __m128d __b)
/// \brief Compares the lower double-precision floating-point values in each of
///    the two 128-bit floating-point vectors of [2 x double] to determine if
///    the value in the first parameter is unequal to the corresponding value in
-///    the second parameter. The comparison yields 0 for false, 1 for true. If
-///    either of the two lower double-precision values is NaN, 0 is returned.
+///    the second parameter.
+///
+///    The comparison yields 0 for false, 1 for true. If either of the two lower
+///    double-precision values is NaN, 0 is returned.
///
///
@@ -1278,8 +1319,9 @@ _mm_cvtps_pd(__m128 __a)

/// \brief Converts the lower two integer elements of a 128-bit vector of
///    [4 x i32] into two double-precision floating-point values, returned in a
-///    128-bit vector of [2 x double]. The upper two elements of the input
-///    vector are unused.
+///    128-bit vector of [2 x double].
+///
+///    The upper two elements of the input vector are unused.
///
///
@@ -1287,7 +1329,9 @@ _mm_cvtps_pd(__m128 __a)
///
/// \param __a
///    A 128-bit integer vector of [4 x i32]. The lower two integer elements are
-///    converted to double-precision values. The upper two elements are unused.
+///    converted to double-precision values.
+///
+///    The upper two elements are unused.
/// \returns A 128-bit vector of [2 x double] containing the converted values.
static __inline__ __m128d __DEFAULT_FN_ATTRS
_mm_cvtepi32_pd(__m128i __a)
@@ -1409,10 +1453,11 @@ _mm_cvtss_sd(__m128d __a, __m128 __b)

/// \brief Converts the two double-precision floating-point elements of a
///    128-bit vector of [2 x double] into two signed 32-bit integer values,
-///    returned in the lower 64 bits of a 128-bit vector of [4 x i32]. If the
-///    result of either conversion is inexact, the result is truncated (rounded
-///    towards zero) regardless of the current MXCSR setting. The upper 64 bits
-///    of the result vector are set to zero.
+///    returned in the lower 64 bits of a 128-bit vector of [4 x i32].
+///
+///    If the result of either conversion is inexact, the result is truncated
+///    (rounded towards zero) regardless of the current MXCSR setting. The upper
+///    64 bits of the result vector are set to zero.
///
///
@@ -1466,9 +1511,10 @@ _mm_cvtpd_pi32(__m128d __a)

/// \brief Converts the two double-precision floating-point elements of a
///    128-bit vector of [2 x double] into two signed 32-bit integer values,
-///    returned in a 64-bit vector of [2 x i32]. If the result of either
-///    conversion is inexact, the result is truncated (rounded towards zero)
-///    regardless of the current MXCSR setting.
+///    returned in a 64-bit vector of [2 x i32].
+///
+///    If the result of either conversion is inexact, the result is truncated
+///    (rounded towards zero) regardless of the current MXCSR setting.
///
///
@@ -1980,8 +2026,9 @@ _mm_storel_pd(double *__dp, __m128d __a)

/// \brief Adds the corresponding elements of two 128-bit vectors of [16 x i8],
///    saving the lower 8 bits of each sum in the corresponding element of a
-///    128-bit result vector of [16 x i8]. The integer elements of both
-///    parameters can be either signed or unsigned.
+///    128-bit result vector of [16 x i8].
+///
+///    The integer elements of both parameters can be either signed or unsigned.
///
///
@@ -2001,8 +2048,9 @@ _mm_add_epi8(__m128i __a, __m128i __b)

/// \brief Adds the corresponding elements of two 128-bit vectors of [8 x i16],
///    saving the lower 16 bits of each sum in the corresponding element of a
-///    128-bit result vector of [8 x i16]. The integer elements of both
-///    parameters can be either signed or unsigned.
+///    128-bit result vector of [8 x i16].
+///
+///    The integer elements of both parameters can be either signed or unsigned.
///
///
@@ -2022,8 +2070,9 @@ _mm_add_epi16(__m128i __a, __m128i __b)

/// \brief Adds the corresponding elements of two 128-bit vectors of [4 x i32],
///    saving the lower 32 bits of each sum in the corresponding element of a
-///    128-bit result vector of [4 x i32]. The integer elements of both
-///    parameters can be either signed or unsigned.
+///    128-bit result vector of [4 x i32].
+///
+///    The integer elements of both parameters can be either signed or unsigned.
///
///
@@ -2061,8 +2110,9 @@ _mm_add_si64(__m64 __a, __m64 __b)

/// \brief Adds the corresponding elements of two 128-bit vectors of [2 x i64],
///    saving the lower 64 bits of each sum in the corresponding element of a
-///    128-bit result vector of [2 x i64]. The integer elements of both
-///    parameters can be either signed or unsigned.
+///    128-bit result vector of [2 x i64].
+///
+///    The integer elements of both parameters can be either signed or unsigned.
///
///
@@ -2208,10 +2258,12 @@ _mm_avg_epu16(__m128i __a, __m128i __b)
/// \brief Multiplies the corresponding elements of two 128-bit signed [8 x i16]
///    vectors, producing eight intermediate 32-bit signed integer products, and
///    adds the consecutive pairs of 32-bit products to form a 128-bit signed
-///    [4 x i32] vector. For example, bits [15:0] of both parameters are
-///    multiplied producing a 32-bit product, bits [31:16] of both parameters
-///    are multiplied producing a 32-bit product, and the sum of those two
-///    products becomes bits [31:0] of the result.
+///    [4 x i32] vector.
+///
+///    For example, bits [15:0] of both parameters are multiplied producing a
+///    32-bit product, bits [31:16] of both parameters are multiplied producing
+///    a 32-bit product, and the sum of those two products becomes bits [31:0]
+///    of the result.
///
///
@@ -3146,8 +3198,9 @@ _mm_cmpgt_epi8(__m128i __a, __m128i __b)

/// \brief Compares each of the corresponding signed 16-bit values of the
///    128-bit integer vectors to determine if the values in the first operand
-///    are greater than those in the second operand. Each comparison yields 0h
-///    for false, FFFFh for true.
+///    are greater than those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFh for true.
///
///
@@ -3166,8 +3219,9 @@ _mm_cmpgt_epi16(__m128i __a, __m128i __b

/// \brief Compares each of the corresponding signed 32-bit values of the
///    128-bit integer vectors to determine if the values in the first operand
-///    are greater than those in the second operand. Each comparison yields 0h
-///    for false, FFFFFFFFh for true.
+///    are greater than those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFFFFFh for true.
///
///
@@ -3186,8 +3240,9 @@ _mm_cmpgt_epi32(__m128i __a, __m128i __b

/// \brief Compares each of the corresponding signed 8-bit values of the 128-bit
///    integer vectors to determine if the values in the first operand are less
-///    than those in the second operand. Each comparison yields 0h for false,
-///    FFh for true.
+///    than those in the second operand.
+///
+///    Each comparison yields 0h for false, FFh for true.
///
///
@@ -3206,8 +3261,9 @@ _mm_cmplt_epi8(__m128i __a, __m128i __b)

/// \brief Compares each of the corresponding signed 16-bit values of the
///    128-bit integer vectors to determine if the values in the first operand
-///    are less than those in the second operand. Each comparison yields 0h for
-///    false, FFFFh for true.
+///    are less than those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFh for true.
///
///
@@ -3226,8 +3282,9 @@ _mm_cmplt_epi16(__m128i __a, __m128i __b

/// \brief Compares each of the corresponding signed 32-bit values of the
///    128-bit integer vectors to determine if the values in the first operand
-///    are less than those in the second operand. Each comparison yields 0h for
-///    false, FFFFFFFFh for true.
+///    are less than those in the second operand.
+///
+///    Each comparison yields 0h for false, FFFFFFFFh for true.
///
///
@@ -3925,10 +3982,11 @@ _mm_storeu_si128(__m128i *__p, __m128i _

/// \brief Moves bytes selected by the mask from the first operand to the
///    specified unaligned memory location. When a mask bit is 1, the
-///    corresponding byte is written, otherwise it is not written. To minimize
-///    caching, the date is flagged as non-temporal (unlikely to be used again
-///    soon). Exception and trap behavior for elements not selected for storage
-///    to memory are implementation dependent.
+///    corresponding byte is written, otherwise it is not written.
+///
+///    To minimize caching, the date is flagged as non-temporal (unlikely to be
+///    used again soon). Exception and trap behavior for elements not selected
+///    for storage to memory are implementation dependent.
///
///
@@ -3972,8 +4030,10 @@ _mm_storel_epi64(__m128i *__p, __m128i _
}

/// \brief Stores a 128-bit floating point vector of [2 x double] to a 128-bit
-///    aligned memory location. To minimize caching, the data is flagged as
-///    non-temporal (unlikely to be used again soon).
+///    aligned memory location.
+///
+///    To minimize caching, the data is flagged as non-temporal (unlikely to be
+///    used again soon).
///
///
@@ -3990,6 +4050,7 @@ _mm_stream_pd(double *__p, __m128d __a)
}

/// \brief Stores a 128-bit integer vector to a 128-bit aligned memory location.
+///
///    To minimize caching, the data is flagged as non-temporal (unlikely to be
///    used again soon).
///
@@ -4007,8 +4068,9 @@ _mm_stream_si128(__m128i *__p, __m128i _
__builtin_nontemporal_store((__v2di)__a, (__v2di*)__p);
}

-/// \brief Stores a 32-bit integer value in the specified memory location. To
-///    minimize caching, the data is flagged as non-temporal (unlikely to be
+/// \brief Stores a 32-bit integer value in the specified memory location.
+///
+///    To minimize caching, the data is flagged as non-temporal (unlikely to be
///    used again soon).
///
@@ -4026,8 +4088,9 @@ _mm_stream_si32(int *__p, int __a)
}

#ifdef __x86_64__
-/// \brief Stores a 64-bit integer value in the specified memory location. To
-///    minimize caching, the data is flagged as non-temporal (unlikely to be
+/// \brief Stores a 64-bit integer value in the specified memory location.
+///
+///    To minimize caching, the data is flagged as non-temporal (unlikely to be
///    used again soon).
///

==============================================================================
+++ cfe/trunk/lib/Headers/mmintrin.h Sun May 14 22:25:04 2017
@@ -608,10 +608,11 @@ _mm_subs_pi16(__m64 __m1, __m64 __m2)

/// \brief Subtracts each 8-bit unsigned integer element of the second 64-bit
///    integer vector of [8 x i8] from the corresponding 8-bit unsigned integer
-///    element of the first 64-bit integer vector of [8 x i8]. If an element of
-///    the first vector is less than the corresponding element of the second
-///    vector, the result is saturated to 0. The results are packed into a
-///    64-bit integer vector of [8 x i8].
+///    element of the first 64-bit integer vector of [8 x i8].
+///
+///    If an element of the first vector is less than the corresponding element
+///    of the second vector, the result is saturated to 0. The results are
+///    packed into a 64-bit integer vector of [8 x i8].
///
///
@@ -631,10 +632,11 @@ _mm_subs_pu8(__m64 __m1, __m64 __m2)

/// \brief Subtracts each 16-bit unsigned integer element of the second 64-bit
///    integer vector of [4 x i16] from the corresponding 16-bit unsigned
-///    integer element of the first 64-bit integer vector of [4 x i16]. If an
-///    element of the first vector is less than the corresponding element of the
-///    second vector, the result is saturated to 0. The results are packed into
-///    a 64-bit integer vector of [4 x i16].
+///    integer element of the first 64-bit integer vector of [4 x i16].
+///
+///    If an element of the first vector is less than the corresponding element
+///    of the second vector, the result is saturated to 0. The results are
+///    packed into a 64-bit integer vector of [4 x i16].
///
///
@@ -657,9 +659,11 @@ _mm_subs_pu16(__m64 __m1, __m64 __m2)
///    element of the second 64-bit integer vector of [4 x i16] and get four
///    32-bit products. Adds adjacent pairs of products to get two 32-bit sums.
///    The lower 32 bits of these two sums are packed into a 64-bit integer
-///    vector of [2 x i32]. For example, bits [15:0] of both parameters are
-///    multiplied, bits [31:16] of both parameters are multiplied, and the sum
-///    of both results is written to bits [31:0] of the result.
+///    vector of [2 x i32].
+///
+///    For example, bits [15:0] of both parameters are multiplied, bits [31:16]
+///    of both parameters are multiplied, and the sum of both results is written
+///    to bits [31:0] of the result.
///
///
@@ -851,10 +855,11 @@ _mm_slli_si64(__m64 __m, int __count)

/// \brief Right-shifts each 16-bit integer element of the first parameter,
///    which is a 64-bit integer vector of [4 x i16], by the number of bits
-///    specified by the second parameter, which is a 64-bit integer. High-order
-///    bits are filled with the sign bit of the initial value of each 16-bit
-///    element. The 16-bit results are packed into a 64-bit integer vector of
-///    [4 x i16].
+///    specified by the second parameter, which is a 64-bit integer.
+///
+///    High-order bits are filled with the sign bit of the initial value of each
+///    16-bit element. The 16-bit results are packed into a 64-bit integer
+///    vector of [4 x i16].
///
///
@@ -874,6 +879,7 @@ _mm_sra_pi16(__m64 __m, __m64 __count)

/// \brief Right-shifts each 16-bit integer element of a 64-bit integer vector
///    of [4 x i16] by the number of bits specified by a 32-bit integer.
+///
///    High-order bits are filled with the sign bit of the initial value of each
///    16-bit element. The 16-bit results are packed into a 64-bit integer
///    vector of [4 x i16].
@@ -896,10 +902,11 @@ _mm_srai_pi16(__m64 __m, int __count)

/// \brief Right-shifts each 32-bit integer element of the first parameter,
///    which is a 64-bit integer vector of [2 x i32], by the number of bits
-///    specified by the second parameter, which is a 64-bit integer. High-order
-///    bits are filled with the sign bit of the initial value of each 32-bit
-///    element. The 32-bit results are packed into a 64-bit integer vector of
-///    [2 x i32].
+///    specified by the second parameter, which is a 64-bit integer.
+///
+///    High-order bits are filled with the sign bit of the initial value of each
+///    32-bit element. The 32-bit results are packed into a 64-bit integer
+///    vector of [2 x i32].
///
///
@@ -919,6 +926,7 @@ _mm_sra_pi32(__m64 __m, __m64 __count)

/// \brief Right-shifts each 32-bit integer element of a 64-bit integer vector
///    of [2 x i32] by the number of bits specified by a 32-bit integer.
+///
///    High-order bits are filled with the sign bit of the initial value of each
///    32-bit element. The 32-bit results are packed into a 64-bit integer
///    vector of [2 x i32].
@@ -941,9 +949,10 @@ _mm_srai_pi32(__m64 __m, int __count)

/// \brief Right-shifts each 16-bit integer element of the first parameter,
///    which is a 64-bit integer vector of [4 x i16], by the number of bits
-///    specified by the second parameter, which is a 64-bit integer. High-order
-///    bits are cleared. The 16-bit results are packed into a 64-bit integer
-///    vector of [4 x i16].
+///    specified by the second parameter, which is a 64-bit integer.
+///
+///    High-order bits are cleared. The 16-bit results are packed into a 64-bit
+///    integer vector of [4 x i16].
///
///
@@ -963,6 +972,7 @@ _mm_srl_pi16(__m64 __m, __m64 __count)

/// \brief Right-shifts each 16-bit integer element of a 64-bit integer vector
///    of [4 x i16] by the number of bits specified by a 32-bit integer.
+///
///    High-order bits are cleared. The 16-bit results are packed into a 64-bit
///    integer vector of [4 x i16].
///
@@ -984,9 +994,10 @@ _mm_srli_pi16(__m64 __m, int __count)

/// \brief Right-shifts each 32-bit integer element of the first parameter,
///    which is a 64-bit integer vector of [2 x i32], by the number of bits
-///    specified by the second parameter, which is a 64-bit integer. High-order
-///    bits are cleared. The 32-bit results are packed into a 64-bit integer
-///    vector of [2 x i32].
+///    specified by the second parameter, which is a 64-bit integer.
+///
+///    High-order bits are cleared. The 32-bit results are packed into a 64-bit
+///    integer vector of [2 x i32].
///
///
@@ -1006,6 +1017,7 @@ _mm_srl_pi32(__m64 __m, __m64 __count)

/// \brief Right-shifts each 32-bit integer element of a 64-bit integer vector
///    of [2 x i32] by the number of bits specified by a 32-bit integer.
+///
///    High-order bits are cleared. The 32-bit results are packed into a 64-bit
///    integer vector of [2 x i32].
///
@@ -1026,8 +1038,9 @@ _mm_srli_pi32(__m64 __m, int __count)
}

/// \brief Right-shifts the first 64-bit integer parameter by the number of bits
-///    specified by the second 64-bit integer parameter. High-order bits are
-///    cleared.
+///    specified by the second 64-bit integer parameter.
+///
+///    High-order bits are cleared.
///
///
@@ -1046,7 +1059,9 @@ _mm_srl_si64(__m64 __m, __m64 __count)

/// \brief Right-shifts the first parameter, which is a 64-bit integer, by the
///    number of bits specified by the second parameter, which is a 32-bit
-///    integer. High-order bits are cleared.
+///    integer.
+///
+///    High-order bits are cleared.
///
///
@@ -1140,8 +1155,9 @@ _mm_xor_si64(__m64 __m1, __m64 __m2)

/// \brief Compares the 8-bit integer elements of two 64-bit integer vectors of
///    [8 x i8] to determine if the element of the first vector is equal to the
-///    corresponding element of the second vector. The comparison yields 0 for
-///    false, 0xFF for true.
+///    corresponding element of the second vector.
+///
+///    The comparison yields 0 for false, 0xFF for true.
///
///
@@ -1161,8 +1177,9 @@ _mm_cmpeq_pi8(__m64 __m1, __m64 __m2)

/// \brief Compares the 16-bit integer elements of two 64-bit integer vectors of
///    [4 x i16] to determine if the element of the first vector is equal to the
-///    corresponding element of the second vector. The comparison yields 0 for
-///    false, 0xFFFF for true.
+///    corresponding element of the second vector.
+///
+///    The comparison yields 0 for false, 0xFFFF for true.
///
///
@@ -1182,8 +1199,9 @@ _mm_cmpeq_pi16(__m64 __m1, __m64 __m2)

/// \brief Compares the 32-bit integer elements of two 64-bit integer vectors of
///    [2 x i32] to determine if the element of the first vector is equal to the
-///    corresponding element of the second vector. The comparison yields 0 for
-///    false, 0xFFFFFFFF for true.
+///    corresponding element of the second vector.
+///
+///    The comparison yields 0 for false, 0xFFFFFFFF for true.
///
///
@@ -1203,8 +1221,9 @@ _mm_cmpeq_pi32(__m64 __m1, __m64 __m2)

/// \brief Compares the 8-bit integer elements of two 64-bit integer vectors of
///    [8 x i8] to determine if the element of the first vector is greater than
-///    the corresponding element of the second vector. The comparison yields 0
-///    for false, 0xFF for true.
+///    the corresponding element of the second vector.
+///
+///    The comparison yields 0 for false, 0xFF for true.
///
///
@@ -1224,8 +1243,9 @@ _mm_cmpgt_pi8(__m64 __m1, __m64 __m2)

/// \brief Compares the 16-bit integer elements of two 64-bit integer vectors of
///    [4 x i16] to determine if the element of the first vector is greater than
-///    the corresponding element of the second vector. The comparison yields 0
-///    for false, 0xFFFF for true.
+///    the corresponding element of the second vector.
+///
+///    The comparison yields 0 for false, 0xFFFF for true.
///
///
@@ -1245,8 +1265,9 @@ _mm_cmpgt_pi16(__m64 __m1, __m64 __m2)

/// \brief Compares the 32-bit integer elements of two 64-bit integer vectors of
///    [2 x i32] to determine if the element of the first vector is greater than
-///    the corresponding element of the second vector. The comparison yields 0
-///    for false, 0xFFFFFFFF for true.
+///    the corresponding element of the second vector.
+///
+///    The comparison yields 0 for false, 0xFFFFFFFF for true.
///
///

==============================================================================
+++ cfe/trunk/lib/Headers/pmmintrin.h Sun May 14 22:25:04 2017
@@ -31,9 +31,11 @@
__attribute__((__always_inline__, __nodebug__, __target__("sse3")))

/// \brief Loads data from an unaligned memory location to elements in a 128-bit
-///    vector. If the address of the data is not 16-byte aligned, the
-///    instruction may read two adjacent aligned blocks of memory to retrieve
-///    the requested data.
+///    vector.
+///
+///    If the address of the data is not 16-byte aligned, the instruction may
+///    read two adjacent aligned blocks of memory to retrieve the requested
+///    data.
///
///

==============================================================================
+++ cfe/trunk/lib/Headers/prfchwintrin.h Sun May 14 22:25:04 2017
@@ -50,8 +50,10 @@ _m_prefetch(void *__P)
///    the L1 data cache and sets the cache-coherency to modified. This
///    provides a hint to the processor that the cache line will be modified.
///    It is intended for use when the cache line will be written to shortly
-///    after the prefetch is performed. Note that the effect of this intrinsic
-///    is dependent on the processor implementation.
+///    after the prefetch is performed.
+///
+///    Note that the effect of this intrinsic is dependent on the processor
+///    implementation.
///
///

==============================================================================
+++ cfe/trunk/lib/Headers/smmintrin.h Sun May 14 22:25:04 2017
@@ -586,7 +586,9 @@ _mm_mul_epi32 (__m128i __V1, __m128i __V
/* SSE4 Floating Point Dot Product Instructions.  */
/// \brief Computes the dot product of the two 128-bit vectors of [4 x float]
///    and returns it in the elements of the 128-bit result vector of
-///    [4 x float]. The immediate integer operand controls which input elements
+///    [4 x float].
+///
+///    The immediate integer operand controls which input elements
///    will contribute to the dot product, and where the final results are
///    returned.
///
@@ -620,7 +622,9 @@ _mm_mul_epi32 (__m128i __V1, __m128i __V

/// \brief Computes the dot product of the two 128-bit vectors of [2 x double]
///    and returns it in the elements of the 128-bit result vector of
-///    [2 x double]. The immediate integer operand controls which input
+///    [2 x double].
+///
+///    The immediate integer operand controls which input
///    elements will contribute to the dot product, and where the final results
///    are returned.
///
@@ -875,7 +879,7 @@ _mm_max_epu32 (__m128i __V1, __m128i __V
/// int _mm_extract_ps(__m128 X, const int N);
/// \endcode
///
-/// This intrinsic corresponds to the <c> VEXTRACTPS / EXTRACTPS </c>
+/// This intrinsic corresponds to the <c> VEXTRACTPS / EXTRACTPS </c>
/// instruction.
///
/// \param X

==============================================================================
+++ cfe/trunk/lib/Headers/tmmintrin.h Sun May 14 22:25:04 2017
@@ -469,10 +469,11 @@ _mm_hsubs_pi16(__m64 __a, __m64 __b)
///    values contained in the first source operand and packed 8-bit signed
///    integer values contained in the second source operand, adds pairs of
///    contiguous products with signed saturation, and writes the 16-bit sums to
-///    the corresponding bits in the destination. For example, bits [7:0] of
-///    both operands are multiplied, bits [15:8] of both operands are
-///    multiplied, and the sum of both results is written to bits [15:0] of the
-///    destination.
+///    the corresponding bits in the destination.
+///
+///    For example, bits [7:0] of both operands are multiplied, bits [15:8] of
+///    both operands are multiplied, and the sum of both results is written to
+///    bits [15:0] of the destination.
///
///
@@ -502,10 +503,11 @@ _mm_maddubs_epi16(__m128i __a, __m128i _
///    values contained in the first source operand and packed 8-bit signed
///    integer values contained in the second source operand, adds pairs of
///    contiguous products with signed saturation, and writes the 16-bit sums to
-///    the corresponding bits in the destination. For example, bits [7:0] of
-///    both operands are multiplied, bits [15:8] of both operands are
-///    multiplied, and the sum of both results is written to bits [15:0] of the
-///    destination.
+///    the corresponding bits in the destination.
+///
+///    For example, bits [7:0] of both operands are multiplied, bits [15:8] of
+///    both operands are multiplied, and the sum of both results is written to
+///    bits [15:0] of the destination.
///
///
@@ -619,13 +621,14 @@ _mm_shuffle_pi8(__m64 __a, __m64 __b)
}

/// \brief For each 8-bit integer in the first source operand, perform one of
-///    the following actions as specified by the second source operand: If the
-///    byte in the second source is negative, calculate the two's complement of
-///    the corresponding byte in the first source, and write that value to the
-///    destination. If the byte in the second source is positive, copy the
-///    corresponding byte from the first source to the destination. If the byte
-///    in the second source is zero, clear the corresponding byte in the
-///    destination.
+///    the following actions as specified by the second source operand.
+///
+///    If the byte in the second source is negative, calculate the two's
+///    complement of the corresponding byte in the first source, and write that
+///    value to the destination. If the byte in the second source is positive,
+///    copy the corresponding byte from the first source to the destination. If
+///    the byte in the second source is zero, clear the corresponding byte in
+///    the destination.
///
///
@@ -644,13 +647,14 @@ _mm_sign_epi8(__m128i __a, __m128i __b)
}

/// \brief For each 16-bit integer in the first source operand, perform one of
-///    the following actions as specified by the second source operand: If the
-///    word in the second source is negative, calculate the two's complement of
-///    the corresponding word in the first source, and write that value to the
-///    destination. If the word in the second source is positive, copy the
-///    corresponding word from the first source to the destination. If the word
-///    in the second source is zero, clear the corresponding word in the
-///    destination.
+///    the following actions as specified by the second source operand.
+///
+///    If the word in the second source is negative, calculate the two's
+///    complement of the corresponding word in the first source, and write that
+///    value to the destination. If the word in the second source is positive,
+///    copy the corresponding word from the first source to the destination. If
+///    the word in the second source is zero, clear the corresponding word in
+///    the destination.
///
///
@@ -669,8 +673,9 @@ _mm_sign_epi16(__m128i __a, __m128i __b)
}

/// \brief For each 32-bit integer in the first source operand, perform one of
-///    the following actions as specified by the second source operand: If the
-///    doubleword in the second source is negative, calculate the two's
+///    the following actions as specified by the second source operand.
+///
+///    If the doubleword in the second source is negative, calculate the two's
///    complement of the corresponding word in the first source, and write that
///    value to the destination. If the doubleword in the second source is
///    positive, copy the corresponding word from the first source to the
@@ -694,13 +699,14 @@ _mm_sign_epi32(__m128i __a, __m128i __b)
}

/// \brief For each 8-bit integer in the first source operand, perform one of
-///    the following actions as specified by the second source operand: If the
-///    byte in the second source is negative, calculate the two's complement of
-///    the corresponding byte in the first source, and write that value to the
-///    destination. If the byte in the second source is positive, copy the
-///    corresponding byte from the first source to the destination. If the byte
-///    in the second source is zero, clear the corresponding byte in the
-///    destination.
+///    the following actions as specified by the second source operand.
+///
+///    If the byte in the second source is negative, calculate the two's
+///    complement of the corresponding byte in the first source, and write that
+///    value to the destination. If the byte in the second source is positive,
+///    copy the corresponding byte from the first source to the destination. If
+///    the byte in the second source is zero, clear the corresponding byte in
+///    the destination.
///
///
@@ -719,13 +725,14 @@ _mm_sign_pi8(__m64 __a, __m64 __b)
}

/// \brief For each 16-bit integer in the first source operand, perform one of
-///    the following actions as specified by the second source operand: If the
-///    word in the second source is negative, calculate the two's complement of
-///    the corresponding word in the first source, and write that value to the
-///    destination. If the word in the second source is positive, copy the
-///    corresponding word from the first source to the destination. If the word
-///    in the second source is zero, clear the corresponding word in the
-///    destination.
+///    the following actions as specified by the second source operand.
+///
+///    If the word in the second source is negative, calculate the two's
+///    complement of the corresponding word in the first source, and write that
+///    value to the destination. If the word in the second source is positive,
+///    copy the corresponding word from the first source to the destination. If
+///    the word in the second source is zero, clear the corresponding word in
+///    the destination.
///
///
@@ -744,8 +751,9 @@ _mm_sign_pi16(__m64 __a, __m64 __b)
}

/// \brief For each 32-bit integer in the first source operand, perform one of
-///    the following actions as specified by the second source operand: If the
-///    doubleword in the second source is negative, calculate the two's
+///    the following actions as specified by the second source operand.
+///
+///    If the doubleword in the second source is negative, calculate the two's
///    complement of the corresponding doubleword in the first source, and
///    write that value to the destination. If the doubleword in the second
///    source is positive, copy the corresponding doubleword from the first

==============================================================================
+++ cfe/trunk/lib/Headers/xmmintrin.h Sun May 14 22:25:04 2017
@@ -2331,8 +2331,10 @@ _mm_mulhi_pu16(__m64 __a, __m64 __b)
/// \brief Conditionally copies the values from each 8-bit element in the first
///    64-bit integer vector operand to the specified memory location, as
///    specified by the most significant bit in the corresponding element in the
-///    second 64-bit integer vector operand. To minimize caching, the data is
-///    flagged as non-temporal (unlikely to be used again soon).
+///    second 64-bit integer vector operand.
+///
+///    To minimize caching, the data is flagged as non-temporal
+///    (unlikely to be used again soon).
///
///
@@ -2815,11 +2817,12 @@ _mm_cvtpi32x2_ps(__m64 __a, __m64 __b)

/// \brief Converts each single-precision floating-point element of a 128-bit
///    floating-point vector of [4 x float] into a 16-bit signed integer, and
-///    packs the results into a 64-bit integer vector of [4 x i16]. If the
-///    floating-point element is NaN or infinity, or if the floating-point
-///    element is greater than 0x7FFFFFFF or less than -0x8000, it is converted
-///    to 0x8000. Otherwise if the floating-point element is greater than
-///    0x7FFF, it is converted to 0x7FFF.
+///    packs the results into a 64-bit integer vector of [4 x i16].
+///
+///    If the floating-point element is NaN or infinity, or if the
+///    floating-point element is greater than 0x7FFFFFFF or less than -0x8000,
+///    it is converted to 0x8000. Otherwise if the floating-point element is
+///    greater than 0x7FFF, it is converted to 0x7FFF.
///
///
@@ -2845,11 +2848,12 @@ _mm_cvtps_pi16(__m128 __a)
/// \brief Converts each single-precision floating-point element of a 128-bit
///    floating-point vector of [4 x float] into an 8-bit signed integer, and
///    packs the results into the lower 32 bits of a 64-bit integer vector of
-///    [8 x i8]. The upper 32 bits of the vector are set to 0. If the
-///    floating-point element is NaN or infinity, or if the floating-point
-///    element is greater than 0x7FFFFFFF or less than -0x80, it is converted
-///    to 0x80. Otherwise if the floating-point element is greater than 0x7F,
-///    it is converted to 0x7F.
+///    [8 x i8]. The upper 32 bits of the vector are set to 0.
+///
+///    If the floating-point element is NaN or infinity, or if the
+///    floating-point element is greater than 0x7FFFFFFF or less than -0x80, it
+///    is converted to 0x80. Otherwise if the floating-point element is greater
+///    than 0x7F, it is converted to 0x7F.
///